The temporal relationship between the solar wind and magnetospheric activity has been studied using 34 intervals of high time resolution IMP 8 solar wind data and the corresponding AL auroral activity index. The median value of the AL index for each interval were utilized to rank the intervals according to geomagnetic activity level. The linear prediction filtering technique was then applied to model magnetospheric response as measured by the AL index to the solar wind input function VBs. The linear prediction filtering routine produces a filter of time-lagged response coefficients which estimates the most general linear relationship between the chosen input and output parameters of the magnetospheric system. It is found that the filters are composed of two response pulses speaking at time lags of 20 and 60 min. The amplitude of the 60-min pulse is the larger for moderate activity levels, while the 20-min pulse is the larger for strong activity levels. A possible interpretation is that the 20-min pulse represents magnetospheric activity driven directly by solar wind coupling and that the 60-min pulse represents magnetospheric activity driven by the relase of energy previously stored in the magnetotal. If this interpretation is correct, the linear filtering results suggest that both the driven and the unloading models of magnetospheric response are important facets of a more comprehensive response model.